Amy Abernethy attended Duke's TIP summer program in the second cohort of its residential program.
“I was a part of the second cohort ever to go to TIP as a part of their residential program.”
Person
This page collects what Amy Abernethy has said on two podcasts: The Long Run with Luke Timmerman, and Raising Health from a16z. They cover her career in cancer research, health data and technology, and how the FDA could regulate artificial intelligence and real-world evidence.
On The Long Run, Abernethy described working in Australia from 1999 to 2003 and launching the Duke Cancer Care Research Program in 2004. She also discussed her board seat at athenahealth and her work with Flatiron. On Raising Health, she said she worked on infrastructure so companies could send data straight to the FDA. She also described Verily's goal of unlocking precision health for everyone.
2 episodes · 36 quoted claims
Amy Abernethy attended Duke's TIP summer program in the second cohort of its residential program.
“I was a part of the second cohort ever to go to TIP as a part of their residential program.”
Amy Abernethy's mother wrote a nursing textbook called Dosage Calculations, and Abernethy helped edit it and do its math problems as a child.
“my mom had written a textbook for nurses called Dosage Calculations”
The Duke Clinical Research Institute ran large-scale clinical trials that included sites in Australia, which Abernethy used to learn clinical trial development.
“at that time, the DCRI had large-scale clinical trials, including sites in Australia.”
Amy Abernethy worked in Australia from 1999 to 2003 as a U.S. physician and researcher.
“during this period of time, which was 1999 to 2003, I was there as a rogue U.S. physician in Australia”
Abernethy's Australian cancer pain trial built its data set from Medicare, hospital systems and insurers in Australia.
“we sourced the data set for the trial from the data sets available from Medicare in Australia, from the hospital systems in Australia, from the insurers in Australia.”
The factorial randomized trial of three cancer pain interventions was deployed across the state of South Australia.
“we deployed those different interventions at scale across the state of South Australia.”
Starting in 2004, Amy Abernethy launched the Duke Cancer Care Research Program, initially just a sticker on her door and business cards.
“starting in 2004, I became the Duke Cancer Care Research Program, which was nothing more than a sticker on my door and a few business cards.”
Around 2011-2012, Abernethy, Rob Califf and another colleague started the Center for Learning Healthcare at Duke, an incubator for using data and technology to speed research.
“together with Rob Califf, who subsequently becomes the commissioner of the FDA, and Victor Zhao, who's currently the president of the National Academy of Medicine, we started something called the Center for Learning Healthcare at Duke.”
In 2009 Abernethy wrote a white paper on a rapid learning system for cancer.
“I had written a treatise in 2009 about this idea of what we call a rapid learning system for cancer.”
Abernethy's white paper became CancerLink, which the American Society of Clinical Oncology funded.
“That white paper ultimately became something with an oncology that's known as CancerLink. It got funded by the American Society of Clinical Oncology”
In 2009 Abernethy's group received a National Cancer Institute grant to extract data from unstructured documents using natural language processing and other techniques.
“In 2009, we had a grant from the National Cancer Institute to try and figure out how to solve for this.”
Abernethy argues commercial and venture investors were not ready for this kind of health data work until around 2011-2012, when enough electronic health records existed.
“I would argue that the private or commercial side, even the venture side, was not ready for all of this until circa 2011, 2012”
Amy Abernethy sits on the board of directors of athenahealth, which runs a cloud-based electronic health record used across the U.S.
“So I sit on the board of directors of Athena Health.”
About half of Flatiron's employees work on its software for oncologists and the customers who use it.
“half of our company, really literally half of the FTEs, are focused on building that software and serving that customer base.”
Abernethy says Epic's electronic health record systems often cost over $100 million and sometimes multiple billions, which is too big for small community oncology practices.
“So, you know, Epic is often a hundred plus million dollar proposition, sometimes up to multiple billion.”
Flatiron has created internal structures to wall off teams so confidential client information is not disclosed to Roche or other clients.
“We cannot disclose any of that information. In fact, have created internal structures to wall off the teams”
Amy Abernethy says the COVID-19 genome was sequenced and published for scientists worldwide within 24 hours, which enabled rapid mRNA vaccine development.
“essentially sequencing the genome, publishing it for scientists all around the world to use within 24 hours, and then moving from there.”
Amy Abernethy says that testing the mRNA vaccines used clinical trial methods from around 1995, with little innovation in trial design.
“we hit the wall of evidence generation where we went back to circa 1995 clinical trials to actually try and understand which of the mRNA vaccines worked for whom and when.”
Amy Abernethy says data from Israel's health systems was used to study how long COVID-19 vaccine effectiveness lasts, safety signals and the need for boosters.
“we leveraged, for example, data from the Israel health systems to understand duration of effectiveness and safety signals and do you need a booster.”
Amy Abernethy says the Evidence Accelerator brought the FDA and innovators together so regulators could say which evidence they trusted and what they wanted to know more about.
“We called this the evidence accelerator. And regulators and others could say, I trust this, I don't trust that, here's what I want to know more about.”
Amy Abernethy says Friends of Cancer Research spread real-world data thinking across the industry so that many innovators, not one company, could contribute.
“Friends of Cancer Research promulgated that thinking across the industry so that it wasn't just any one company, but rather all innovators having the chance to say what they thought.”
Amy Abernethy says the FDA issued three core guidances on real-world data at the end of 2021, informed by the collaborative process she described.
“That then informed the guidance that came out from FDA at the end of 2021. There were three core guidances around robot data, all informed by that process.”
Amy Abernethy says using AI to identify drug targets creates little regulatory worry because the work happens before a molecule is put into people.
“there's some parts of the application of AI in the healthcare and life sciences landscape that don't really create a bunch of angst. Identification of targets is a good example of that.”
Amy Abernethy says her first real job, in 1982, was in an artificial intelligence lab at NASA.
“That was where my first real job was in 1982. And so that's a long time ago.”
Amy Abernethy says the visible surge in artificial intelligence development happened at the end of 2022.
“the pace of development in artificial intelligence ultimately happened at the end of 2022”
Amy Abernethy says AI-based products can be regulated under the FDA's traditional device framework, such as diagnostics or software as a medical device.
“We can regulate them according to the traditional device framework, whether that's a diagnostic or a software as a medical device.”
Amy Abernethy says the FDA published an action plan for AI-based medical software around January 2021.
“And I think it was like around January 2021, put out an action plan to specify, here are the details of what we think good's going to look like in this space.”
Amy Abernethy says the FDA's AI action plan called for good software design and an easy way to understand an AI product's output.
“We need, you know, good software design. We need to make sure that there is an easy way to understand the output of an AI-based product”
Amy Abernethy says the FDA's AI action plan also called for validation testing and performance monitoring over time.
“we need appropriate validation testing and we need some mechanism for performance monitoring across time.”
Amy Abernethy says that, as far as she knew, the FDA's AI action plan had not been turned into a new rule set.
“that actually hasn't translated to a new rule set, as far as I know, beyond what we've previously had.”
Amy Abernethy says companies traditionally submit clinical trial results to the FDA as hundreds of pages of reports and then wait for questions.
“you ultimately package up the clinical trial, write hundreds of pages of reports, and ultimately ship that over the FDA and wait for questions to come back.”
Amy Abernethy says that while at the FDA she worked on infrastructure so companies could push data directly to the agency instead of submitting lengthy reports.
“we were building the pipes at FDA to be ready for a landscape where really all you do is push your data across over FDA and FDA is ready to receive”
Amy Abernethy proposes that a foundation model for reading chest X-rays could be regulated once, so specialized models built on it need less review.
“a foundational model of reading a chest X-ray. And then that's already done, a check for the foundational model.”
Amy Abernethy says statistical programs written 10 to 15 years ago began targeting clinical trial monitors' work at higher-risk sites, which became risk-based monitoring.
“pattern recognition and statistical programs were written about 10, 15 years ago to start to do that and better target the work of monitors.”
Amy Abernethy says Verily's goal is to unlock the promise of precision health for everyone every day.
“our goal is to unlock the promise of precision health for everyone every day.”
Amy Abernethy says Verily builds software for clinical research sites, clinical trials and patient recruitment, plus products in healthcare delivery such as diabetes management.
“software for clinical research sites, software to conduct clinical trials, patient recruitment, as well as products in the healthcare delivery ecosystem”